EFFECTS OF INBEEBDING AND CROSSBREEDING. 39 



The genetic differentiation among the famihes is obvious. It is 

 further interesting to note that there was segregation in this respect 

 among the lines of descent within the famihes. 



The history of Family 31 is especially interesting. The original 

 pair were both normal as to toes and produced only normal young. 

 Two matings were made in the next generation, neither of which 

 produced any four-toed young. Sixty-six matings were made among 

 the descendants of one of them. Among the numerous young only 

 2 are recorded as four-toed. One of these came in the fourth gene- 

 ration, the other in the twelfth. The two matings which produced 

 the four-toed young also produced 1 1 normal young. Evidently the 

 vestigial toe was almost below the threshold of appearance in this 

 line. A mating of 2 normals derived from the other mating in the 

 first generation produced 1 1 normals but also 4 f our-toed^ young. 

 Two matings made in the next generation left numerous descendants. 

 The two lines derived from these two matings and thus diverging in 

 the third generation were remarkably different from each other in 

 respect to heredity of polydactylism. One line starting from a 

 mating of normal with normal contained 18 matings, of which only 

 5 produced four-toed young. These 5 matings produced 23 normal 

 young and 6 four-toed young. The other line, starting from a mating 

 of 2 four-toed animals, contained 27 matings, of which 25 produced 

 139 four-toed to 70 normal young. A large percentage of the four- 

 toed young in this line had four well-developed toes on both hind feet. 



There are similar evidences of segregation within Family 35. In 

 this case three out of four lines starting from matings in the second 

 generation produced numerous four-toed young. The other line, 

 containing 43 matings, produced none. The segregation among the 

 family lines is so sharp that it is probable that a careful investigation 

 of polydactylism would yield Mendelian results, though much non- 

 genetic variation must be present. The differentiation among the 

 families and among the lines of descent within families is obviously 

 very similar to that found in the case of the color characters. 



Of more interest from the standpoint of inbreeding are abnormali- 

 ties which are not merely reappearances of formerly normal characters. 

 It has often been held that inbreeding has a specific tendency to 

 cause physical malformations to appear. The array of bottles of 

 preserved specimens of abnormalities which have appeared in the 

 present experiments is indeed rather imposing. The total number 

 of young born, however, has been so great that the sum of all of the 

 abnormalities (excluding polydactylism) forms really an insignificant 

 proportion. 



Most of the types of abnormalities have appeared also in the con- 

 trol stock, showing that inbreeding can not be the prime cause of their 

 appearance. Moreover, a study of their distribution among the fami- 



